search

microsoft / MASS

MASS: Masked Sequence to Sequence Pre-training for Language Generation
https://arxiv.org/pdf/1905.02450.pdf
Other
1.11k stars 206 forks source link

What is the difference between pretrain-tensor2tensor and MASS? #156

Closed guotong1988 closed 3 years ago

guotong1988 commented 4 years ago

Thank you very much!

StillKeepTry commented 4 years ago

What is pretrain-tensor2tensor?

guotong1988 commented 4 years ago

https://github.com/tensorflow/tensor2tensor

guotong1988 commented 4 years ago

What is the difference between BART and MASS?

bozhenhhu commented 3 years ago

I guess you may be need to read these paper multi-times to unsderstand their differences

guotong1988 commented 3 years ago

Close because no friendly response.